VU Research Portal

(1)

Biological risk factors and sleep-related risk factors of type 2 diabetes mellitus

Jansen-Koopman, A.D.M.

2019

document version

Publisher's PDF, also known as Version of record

Link to publication in VU Research Portal

citation for published version (APA)

Jansen-Koopman, A. D. M. (2019). Biological risk factors and sleep-related risk factors of type 2 diabetes mellitus: To refine characterization of high risk subjects and to identify novel risk factors.

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal ?

Take down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

E-mail address:

(2)

Chapter 6 The association between multiple sleep-related characteristics

and the metabolic syndrome in the general population: the

New Hoorn study.

Kaira C. van der Pal, Anitra D.M. Koopman, Jeroen Lakerveld, Amber A.W.A.

van der Heijden, Petra J. Elders, Joline W. Beulens, Femke Rutters

(3)

92

ABSTRACT

Background: Previous studies have investigated the association between sleep duration,

insomnia, day-time napping and metabolic syndrome individually, but never conjointly and the association with sleep medication use has yet to be investigated. Our aim was to examine the associations between these sleep-related characteristics and the metabolic syndrome in a population-based cohort, individually and conjointly.

Material and methods: We used cross-sectional data of 1679 participants from the New

Hoorn study, 52.6% women and age 60.86.4y. Sleep duration, insomnia and day-time napping were measured using validated questionnaires. Use of sleep medication was measured by registration of dispensing labels. Metabolic syndrome was defined according to ATP III. Linear and Poisson regression were used and all analyses were adjusted for age, sex, education level, job status, smoking, physical activity, depression and BMI.

Results: In our population-based cohort, 447 (26.6%) persons had the metabolic syndrome.

The individual associations showed that after correction, day-time napping for ≤30 minutes and >30 minutes was associated with a prevalence ratio for the metabolic syndrome of 1.28 (95% CI: 1.1-1.5) and 1.74 (95% CI: 1.4-2.2), respectively, compared to participants who did not nap. Sleep duration, insomnia, and sleep medication use were not associated with the metabolic syndrome individually. The conjoint analysis, showed after correction that having ≥2 sleep-related characteristics was associated with a PR of 1.36 (95% CI: 1.0-1.8) of having the metabolic syndrome, compared to having no sleep-related characteristics.

Conclusion: Sleep-related characteristics were associated with a higher prevalence of the

(4)

93

INTRODUCTION

The metabolic syndrome refers to a cluster of metabolic and cardiovascular abnormalities, including visceral obesity, insulin resistance, hypertension and altered lipoprotein profiles [1]. The metabolic syndrome is an increasing problem in developed countries, with recent studies showing that worldwide about a quarter of the population is affected [2]. It is well known that the risk of the metabolic syndrome increases with age, higher Body Mass Index (BMI) as well as the presence of an inactive lifestyle and an unhealthy diet [3]. Besides these frequently studied risk factors, a less obvious risk factor for the metabolic syndrome is sleep. Experimental and observational studies have shown that disturbance of sleep, comprising both duration and quality, increased the risk of metabolic disturbance, the metabolic syndrome and type 2 diabetes mellitus (T2DM) [4–6]. A recent meta-analysis based on data from 12 cross-sectional studies, confirmed that in the general population short sleep duration was associated with an 27% increased risk of having the metabolic syndrome (OR = 1.27, 95% CI: 1.01-1.5) [7].

However, these previous studies were mostly conducted in only men or women and, more importantly, data on the associations between other sleep-related characteristics besides sleep duration and quality, such as insomnia (difficulty falling asleep, waking up in the middle of the night, or self-reported low quality of sleep) and day-time napping, and the metabolic syndrome is scarce. With regards to insomnia, one study has shown a higher risk of the metabolic syndrome in participants that suffer from difficulty falling asleep [8], and two other studies showed such an association with T2DM [9,10]. For day-time napping, two Asian population-based cohort studies have shown a higher prevalence of metabolic syndrome in those who nap [11,12], compared to those who do not. While experimental work suggests effects of fragmented sleep on metabolic parameters [13], more population-based research is needed to show the metabolic effect of these sleep disturbances in the general population.

(5)

day-94

time napping and sleep medication use, and metabolic syndrome in a population-based cohort. The second aim of this study was to examine the joint association of multiple sleep-related characteristic.

MATERIALS AND METHODS

Design and study population

The New Hoorn Study is a population-based cohort. The study was carried out from 2006 to 2007 in the Dutch city of Hoorn, a city of 70.000 inhabitants in the province of West-Friesland. The eligible population of the Hoorn Study consisted of 6180 men and women, aged 40–75 years, who were randomly selected from the municipal registry. Of the eligible participants, 45% agreed to participate, resulting in the New Hoorn Study cohort of 2807 participants.

From 2013-2015, a follow-up examination was performed. Of the original cohort, 2576 participants could be invited. A total of 1815 participants agreed to participate of which 1734 participants actually took part in the follow-up measurements (67%). For the present study, only data from the follow-up measurement were used, since sleep-related characteristics were only assessed at that visit. Participants without information regarding sleep-related characteristics and metabolic syndrome were excluded (n=55). Thus, from the analytical sample consisted of 1679 participants (Figure 1). The Ethics Committee of the VU University Medical Centre approved the New Hoorn Study and written informed consent was obtained from all participants.

Measurements

Participants were asked to refrain from eating and drinking (except drinking water) from 10:00 P.M. the night before the visit and from drinking alcohol from 5:00 P.M. the day before the visit. The participants were asked not to smoke from 10:00 P.M. the night before the visit. Participants who had not been following these instructions were asked to reschedule their visit [19]. Before the visit, participants were asked to complete questionnaires on their medical history, depressive symptoms, diet, physical activity and sleep-related characteristics. During the visit, metabolic syndrome parameters as well as other physical measures were determined.

Sleep-related characteristics

Sleep duration (hours): was assessed by questionnaire by asking the 4 following

(6)

95 of going to bed and getting up on weekdays and weekend days (5*sleep duration week days + 2*sleep duration weekend days/7) [20,21]. Sleep duration was stratified into those who slept less than 7 hours, 7 to less than 8 hours, 8 to less than 9 hours and 9 hours or more. We conjoined the groups with less than 6 hours of sleep and 6 to less than 7 hours, due to the limited number of participants who slept less than 6 hours. Furthermore, we used the group who slept 7 to less than 8 hours on average as the reference group and participants who slept less than 7 hours were defined as short sleepers and those who slept more than 9 hours were defined as long sleepers.

Insomnia (yes/no): Participants were considered to have insomnia if three out of

(7)

96

3) self-reported low quality of sleep, defined as yes or no, 4) scores of ≥10 on the Epworth Sleepiness Scale (ESS).

Day-time napping (min): was self-reported based on the question ‘When possible

do you try to incorporate a nap during the day?' and 'If yes, how many hours/minutes per day do you nap?’. Day-time napping time was stratified into no napping, napping 30 minutes or less or napping more than 30 minutes.

Sleep medication use (yes/no): participants were asked to bring their current

prescribed medication to the visit, which were used to note the prescription labels. The Anatomical Therapeutic Chemical (ATC) Classification System was used for the classification of drugs. Anxiolytics, hypnotics and sedatives were defined as sleep medication, including ATC codes N05B and N05C.

Metabolic syndrome

Metabolic syndrome status was defined according to the ATP III of the National Cholesterol Education Program, i.e. participants had to meet three or more of the following criteria: fasting plasma glucose (FPG) levels ≥ 6.1 mmol/l or the presence of T2DM, high-density lipoprotein (HDL) cholesterol levels < 1.0 mmol/l in men or < 1.3 mmol/l in women, triglyceride levels ≥ 1.7 mmol/l or use of lipid medication, waist circumference ≥ 102 cm in men or ≥ 88 cm in women or ≥ 130/85 mmHg or use of antihypertensive medication [22].

Fasting plasma glucose levels (mmol/l) were determined using the glucose dehydrogenase method (Merck, Darmstad, Germany). Triglycerides and HDL levels (mmol/l) were determined in the fasting blood sample, using enzymatic techniques (Boehringer Mannheim, Mannheim, Germany). Waist circumference (cm) was measured according to a standardised procedure, as described earlier [23]. Sitting systolic and diastolic blood pressure (mmHG) was measured twice on the right arm with a random-zero sphygmomanometer (Hawksley–Gelman, Lancing, UK) and the average was used in the analyses.

Possible confounders and effect modifiers

(8)

97 measured with participants wearing light clothes only, and BMI was calculated as weight/height squared (kg/m2_).

Statistical analysis

Continuous data were presented as mean ± standard deviation (SD), percentages or median and IQR and were log-transformed in case of a skewed distribution. First, participant characteristics stratified for sleep duration were presented. Second, Poisson regression analyses were performed to estimate the prevalence ratios of the metabolic syndrome per sleep-related characteristic, individually. Poisson regression was used here since the metabolic syndrome was a prevalent outcome in our study. Since odds ratios from logistic regression may overestimate the association when the outcome is prevalent, the use of Poisson regression is suggested for these cases [26]. Next, we used linear regression analyses to estimate the beta coefficients per sleep-related characteristic and as a sensitivity analysis per parameter of the metabolic syndrome. Fourth, Poisson regression analyses were performed to estimate the conjoint association of the sleep-related characteristics with the prevalence ratios of the metabolic syndrome, in which we assessed the association of each additional sleep-related characteristic, comparing none versus 1 or

 2 sleep-related characteristics. We defined the sum score by adding having either short sleep duration, long sleep duration, insomnia, day-time napping, and/or sleep medication use as the different sleep-related characteristics. Finally, we calculated the P for trend for the association between additional sleep-related characteristics and metabolic syndrome by using a Chi-Square test.

(9)

98

RESULTS

Of the 1679 participants, 52.6% were women and the average age (SD) was 60.8 (6.4) years. No significant differences were observed with regard to age and metabolic syndrome factors between those with and without sleep information (data not shown). Table 1 shows the participant characteristics stratified for average sleep duration. Participants with the longest sleep duration had the highest percentage of metabolic syndrome, compared to those in the average sleep duration group (7-8 h), namely 30.5% versus 24.5%. With regard to the other sleep-related characteristics, we observed that participants with the shortest average sleep duration had the highest prevalence of insomnia and the longest day-time napping duration and the highest sleep medication use rates.

Table 2 shows the prevalence ratios (PR) of the individual sleep-related characteristics and the metabolic syndrome. We report on the third model, as this is adjusted most completely. Model 3 showed that day-time napping for 30 minutes or less and day-time napping for more than 30 minutes were associated with an increased prevalence of the metabolic syndrome, PR = 1.28 (95% CI: 1.1; 1.5) and 1.74 (95% CI: 1.4; 2.2), respectively, compared to no day-time napping. Including BMI in the model attenuated the association by 7% and 22%, suggesting mediation of the association by BMI. Finally, sleep medication use was associated with a higher prevalence of having the metabolic syndrome in the crude model, PR = 1.41 (95%-CI: 1.0; 1.9). However, this association attenuated to non-significant after multivariable adjustment in model 3, namely PR = 1.21 (95%-CI: 0.8; 1.7).

Supplementary Table 1 shows the beta coefficients of the individual sleep-related characteristics and the parameters of the metabolic syndrome. Participants who reported to nap regularly had significantly higher fasting plasma glucose levels (0.15 mmol/l (95% CI: 0.1;0.3) for those who napped 30 minutes or less, and 0.34 mmol/l (95% CI: 0.2;0.5) for those napping longer), compared to no day-time napping. Short and longer day-time nappers also had a significantly larger waist circumference (1.50 cm (95% CI: -0.3;3.3), and 3.29 cm (95% CI: 0.6;6.0), respectively). Sleep medication use was associated with higher levels of fasting plasma glucose, 0.40 mmol/l (95% CI: 0.2;0.6) as well as log triglyceride levels, 0.14 mmol/l (95% CI: 0.0;0.3), compared to participants who did not use sleep medication.

(10)

N 1679 593 212 635 239 Age (years) 60.8 6.4 59.1 6.4 59.8 6.1 61.8 6.4 62.2 6.3

Sex (% women) 52.6 41.5 47.7 56.4 64.4

Education (%low/middle/high) 6/64/30 9/59/32 4/60/36 5/67/28 12/72/16 Job status (%paid/retired/unemployed) 59/37/4 74/23/3 68/28/4 50/46/5 49/48/3

Metabolic syndrome factors

Fasting plasma glucose (mmol/l) 5.9 0.9 5.9 1.1 5.9 1.0 5.8 0.8 5.9 1.0

HDL (mmol/l) 1.6 0.4 1.6 0.4 1.6 0.5 1.6 0.4 1.5 0.4

Triglycerides (mmol/l) 1.3 [0.9; 1.6] 1.4 [0.9; 1.2] 1.3 [0.8; 1.6] 1.3 [0.9; 1.6] 1.4 [0.9; 1.7] Waist circumference (cm) 91.6 16.0 93.3 12.2 92.4 22.2 90.5 10.7 91.4 12.1 Systolic blood pressure (mmHg) 131 18.8 131 19 130 18 131 19 134 19; Diastolic blood pressure (mmHg) 79 8.4 80 9 79 8 78 8 80 8

Metabolic syndrome (%) 26.6 27.8 24.5 26.8 30.5

Other sleep characteristics

Insomnia (%) 5.2 11.6 6.4 2.8 3.1

Day-time napping (minutes) 13.7 21.2 17.1 25.2 12.4 19.5 12.9 19.8 15.6 24.3

No day-time napping (%) 57.0 53.3 58.5 56.7 57.3

<=30 minutes (%) 31.4 30.2 32.0 32.0 29.3

>30 minutes (%) 11.6 16.5 9.4 11.3 13.4

Sleep medication use (%) 4.5 8.0 2.7 3.6 8.4

Other co-variants and confounders

BMI (kg/m2₎ _26.5__4.2 _27.2__4.6 _26.5__4.2 _26.2__3.9 _26.8__4.6

Depression >= 16 CESD (%) 11.0 14.1 9.0 10.4 15.2

Smokers (%) 58.7 60.7 60.0 56.8 58.9

Sufficient physical activity (%) 65.5 56.0 63.8 70.6 64.3

(11)

100

Table 2. Crude and adjusted prevalence ratios (95% confidence interval) of the metabolic

syndrome for sleep-related characteristics (N=1679).

Model 1 Model 2 Model 3 Model 4

<7 hours 1.14 (0.9; 1.5) 1.14 (0.9; 1.5) 1.04 (0.8; 1.2) 0.96 (0.7; 1.3)

>=7 to <8 hours [Reference] [Reference] [Reference] [Reference]

>=8 to <9 hours 1.10 (0.9; 1.3) 0.99 (0.8; 1.2) 0.99 (0.8; 1.2) 1.07 (0.9; 1.3)

>=9 hours 1.25 (0.99; 1.6) 1.07 (0.8; 1.4) 1.05 (0.8; 1.4) 1.06 (0.8; 1.3)

Insomnia 0.94 (0.6; 1.4) 0.90 (0.6; 1.4) 0.81 (0.5; 1.3) 0.85 (0.6; 1.2)

No nap [Reference] [Reference] [Reference] [Reference]

=<30 minutes 1.27 (1.1; 1.5) 1.30 (1.1; 1.6) 1.28 (1.1; 1.5) 1.19 (1.0; 1.4) >30 minutes 1.76 (1.4; 2.2) 1.86 (1.5; 2.3) 1.74 (1.4; 2.2) 1.38 (1.1; 1.7) Sleep medication 1.41 (1.0; 1.9) 1.34 (0.98; 1.8) 1.21 (0.8; 1.7) 1.25 (0.96; 1.6) Model 1: unadjusted

Model 2: adjusted for age, sex, education, job status

Model 3: as model 2, plus smoking, physical activity, and depression Model 4: as model 2, plus BMI

(12)

101

Table 3. Crude and adjusted prevalence ratios (95% confidence interval) of the metabolic

syndrome for sleep-related characteristics (N=1679).

Metabolic syndrome

Number sleep related measures

N Model 1 Model 2 Model 3 Model 4

0 288 [Reference] [Reference] [Reference] [Reference] 1 733 1.11 (0.9; 1.4) 1.06 (0.8; 1.4) 1.09 (0.8; 1.4) 1.03 (0.8; 1.3) 2-4 487 1.54 (1.2; 2.0) 1.43 (1.1; 1.8) 1.36 (1.0; 1.8) 1.22 (0.9; 1.6) Model 1: unadjusted

Model 3: as model 2, plus smoking, physical activity, and depression Model 4: as model 2, plus BMI

(13)

102

DISCUSSION

We examined the individual and conjoint associations of multiple sleep-related characteristics and the metabolic syndrome in a population-based cohort. We showed that day-time napping was associated with the metabolic syndrome, such that participants who napped 30 minutes or more had a 30% higher chance of having the metabolic syndrome, compared to those who did not nap. The association with day-time napping and the metabolic syndrome was mainly due to the higher fasting glucose concentrations and waist circumference components of the metabolic syndrome. For sleep duration, insomnia, and sleep medication use we did not observe significant associations. Second, we observed a dose-response association, such that two or more sleep-related characteristics were associated with a 40% higher chance of the metabolic syndrome, compared to the absence of sleep-related characteristics.

In contrast with previous studies [7,27], we did not observe a significant association between sleep duration and metabolic syndrome. The majority of previous studies showed 27-62% higher risks of having the metabolic syndrome for short sleep duration [28]. Our observation was in the same direction, but non-significant which might be due to the small amount of participants that slept less than 7 or more than 9 hours in our cohort as well as the correction for multiple confounders. Additionally, we distinguished sleep duration from week and weekend days, while others only measured average sleep duration of the weekdays; this could have resulted in longer sleep duration in our cohort and obscuring the short sleep duration effect. In contrast to previous findings, we also did not observe an association between insomnia and (parameters of) metabolic syndrome [8,10]. These dissimilarities might be explained by the fact that the percentage of participants suffering from insomnia in our cohort was very low, namely only 5.2%. In addition, the difference might also be caused by differences in definition of insomnia, which to date has not been harmonized.

(14)

103 therefore assess if addressing sleep-related characteristics could reduce the risk of the metabolic syndrome can also be focussed on sleep-related characteristics.

Several possible mechanisms could mediate the association between sleep-related characteristics and the metabolic syndrome. First, the hypothalamic-pituitary-adrenal (HPA) axis and sympathetic activation have been suggested to play a role in the development of metabolic syndrome [27]. Hyperactivation of the HPA axis or sympathetic activation leads to an increase in cortisol levels which contribute to insulin resistance, dyslipidemia and abdominal obesity [8,11]. This pathway is line with our results, showing that fasting glucose concentrations, triglycerides and waist circumference were mainly responsible for the association between day-time napping, sleep medication use and an increased risk of the metabolic syndrome, while blood pressure and other blood lipids were not. Second, the association between long sleep duration and metabolic syndrome is mediated by depression [28], as confirmed by our (non-significant) results. Finally, the hunger regulating hormones leptin and ghrelin are associated with sleep parameters. For example, sleep restriction causes an increase in ghrelin levels and a decrease in leptin levels, which regulate satiety and hunger, respectively. This leads to higher intake of food and subsequently weight gain and higher prevalence of metabolic syndrome [7]. Thereby, shorter sleep duration is associated with more sedentary time in contrast to a normal sleep duration [29]. Except for the affected metabolic parameters, our current study does not enable us to assess these other possible pathways and therefore additional studies are needed.

This study has some limitations that need to be mentioned. First, all the sleep data was self-reported, however to date no golden standard is present to measure sleep. Second, we had no information about snoring or sleep apnea, which also could influence the quality of sleep and is strongly associated with the metabolic syndrome. Thirdly, these data were cross-sectional, making it hard to distinguish cause and effect. Fourthly, we had no information on insulin, so no information on insulin resistance and insulin sensitivity if available. Finally, it is not certain if the different sleep-related characteristics are really different from each other or if they all reflect the same issue, namely poor sleeping quality. We believe they reflect different characteristics as the correlations between the variables was minimal (P<0.3 for all). The main strengths of this study were the use of data from a relatively large, middle-aged population-based cohort with detailed phenotyping to allow appropriate adjustment for confounding and we were the first to compare different sleep-related characteristics with each other, instead of looking at a single sleep measure at a time and looking at the effect of the characteristics conjointly.

(15)

Supplementary Table 1. Unstandardized Betas and 95% confidence intervals per sleep disturbances and per parameter of

the metabolic syndrome.

Systolic blood pressure (mmHg) Fasting plasma glucose (mmol/L) High-density lipoprotein (mmol/L) Log triglycerides (mmol/L) Abdominal circumference (cm) Sleep duration model 1

<7 hours 1.09 (-1.9; 4.0) 0.06 (-0.1; 0.2) -0.02 (-0.1; 0.0) 0.06 (0.0; 0.1) 0.90 (-1.7; 3.5)

>=7 to <8 hours [Reference] [Reference] [Reference] [Reference] [Reference]

>=8 to <9 hours 0.45 (-1.7; 2.5) -0.05 (-0.2; 0.0) -0.04 (-0.1; 0.0) 0.03 (0.0; 0.1) -1.86 (-3.7; 0.0)

>=9 hours 3.55 (0.7; 6.4) 0.04 (-0.1; 0.2) -0.06 (-0.1; 0.0) 0.06 (0.0; 0.1) -1.03 (-3.5; 1.4)

Sleep duration model 2

<7 hours 0.95 (-1.9; 3.8) 0.04 (-0.1; 0.2) -0.02 (-0.1; 0.0) 0.05 (0.0; 0.1) 0.15 (-2.3; 2.6)

>=8 to <9 hours -1.32 (-3.4; 0.7) -0.06 (-0.2; 0.0) -0.04 (-0.1; 0.0) 0.03 (0.0; 0.1) -1.57 (-3.4; 0.2)

>=9 hours 1.48 (-1.3; 4.3) 0.04 (-0.1; 0.2) -0.07 (-0.1; 0.0) 0.07 (0.0; 0.1) -0.62 (-3.1; 1.8)

<7 hours 1.34 (-1.6; 4.3) -0.02 (-0.2; 0.1) -0.02 (-0.1; 0.1) 0.02 (-0.1; 0.1) 0.15 (-2.5; 2.8)

>=8 to <9 hours -1.21 (-3.3; 0.9) -0.07 (-0.2; 0.0) -0.03 (-0.1; 0.0) 0.03 (0.0; 0.1) -1.47 (-3.4; 0.4)

>=9 hours 1.84 (-1.1; 4.8) 0.05 (-0.1; 0.2) -0.06 (-0.1; 0.0) 0.06 (0.0; 0.1) -0.56 (-3.2; 2.1)

<7 hours 0.54 (-2.3; 3.3) -0.01 (-0.1; 0.1) -0.02 (-0.1; 0.0) 0.03 (0.0; 0.1) -0.89 (-2.9; 1.1)

(16)

No napping [Reference] [Reference] [Reference] [Reference] [Reference]

<=30 min napping -0.13 (-2.1; 1.9) 0.19 (0.1; 0.3) 0.01 (0.0; 0.1) 0.07 (0.0; 0.1) 2.68 (0.9; 4.4)

>30 min napping -1.17 (-4.1; 1.7) 0.41 (0.3; 0.5) -0.04 (-0.1; 0.0) 0.13 (0.1; 0.2) 4.65 (2.1; 7.2)

Napping model 2

<=30 min napping -0.83 (-2.8; 1.1) 0.15 (0.1; 0.2) 0.01 (0.0; 0.1) 0.05 (0.0; 0.1) 1.44 (-0.3; 3.1)

>30 min napping -1.01 (-3.8-1.8) 0.39 (0.3; 0.5) -0.04 (-0.1; 0.0) 0.12 (0.0; 0.2) 3.80 (1.3; 6.3)

Napping model 3

<=30 min napping -0.94 (-2.9; 1.1) 0.15 (0.1; 0.3) 0.01 (0.0; 0.1) 0.04 (0.0; 0.1) 1.50 (-0.3; 3.3)

>30 min napping -0.65 (-3.7; 2.4) 0.34 (0.2; 0.5) -0.04 (-0.1; 0.0) 0.13 (0.1; 0.2) 3.29 (0.6; 6.0)

Napping model 4

<=30 min napping -1.35 (-3.3; 0.6) 0.10 (0.0; 0.2) 0.01 (0.0; 0.1) 0.03 (0.0; 0.1) 0.01 (-1.4; 1.4) >30 min napping -2.12 (-4.9; 0.7) 0.28 (0.1; 0.4) -0.04 (-0.1; 0.0) 0.06 (0.0; 0.1) 0.34 (-1.7; 2.4) Sleep medication Model 1 -4.17 (-8.5; 0.2) 0.28 (0.1; 0.5) 0.03 (-0.1; 0.1) 0.18 (0.1; 0.3) 0.59 (-3.2; 4.3) Model 2 -3.65 (-7.8; 0.5) 0.34 (0.1; 0.6) 0.02 (-0.1; 0.1) 0.21 (0.1; 0.3) 2.36 (-1.3; 6.0) Model 3 -3.48 (-8.0; 1.0) 0.40 (0.2; 0.6) 0.01 (-0.1; 0.1) 0.14 (0.0; 0.3) 1.81 (-2.2; 5.8) Model 4 -4.14 (-8.3; 0.0) 0.29 (0.1; 0.5) 0.02 (-0.1; 0.1) 0.18 (0.1; 0.3) 0.22 (-2.7; 3.2) Model 1: unadjusted

Model 3: as model 2 plus smoking, physical activity, and depression Model 4: as model 2 plus BMI

(17)

106

REFERENCES

1. Alberti KGMM, Eckel RH, Grundy SM, et al. Harmonizing the metabolic syndrome: a joint

interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World

Heart Federation; International . Circulation. 2009;120(16):1640-1645.

doi:10.1161/CIRCULATIONAHA.109.192644.

2. International Diabetes Federation. http://www.idf.org/. Accessed May 5, 2016.

3. WHO | Health data and statistics. http://www.who.int/healthinfo/statistics/en/. Accessed

May 5, 2016.

4. Shan Z, Ma H, Xie M, et al. Sleep duration and risk of type 2 diabetes: a meta-analysis of

prospective studies. Diabetes Care. 2015;38(3):529-537.

5. Holliday EG, Magee CA, Kritharides L, Banks E, Attia J. Short sleep duration is associated with

risk of future diabetes but not cardiovascular disease: a prospective study and meta-analysis. PLoS One. 2013;8(11):e82305.

6. Cappuccio FP, D’Elia L, Strazzullo P, Miller MA. Quantity and quality of sleep and incidence

of type 2 diabetes: a systematic review and meta-analysis. Diabetes Care. 2010;33(2):414-420.

7. Xi B, He D, Zhang M, Xue J, Zhou D. Short sleep duration predicts risk of metabolic syndrome:

A systematic review and meta-analysis. Sleep Med Rev. 2014;18(4):293-297.

8. Troxel WM, Buysse DJ, Matthews K a, et al. Sleep symptoms predict the development of the

metabolic syndrome. Sleep. 2010;33(12):1633-1640.

9. Knutson KL, Van Cauter E, Zee P, Liu K, Lauderdale DS. Cross-sectional associations between

measures of sleep and markers of glucose metabolism among subjects with and without diabetes: The coronary artery risk development in young adults (CARDIA) sleep study. Diabetes Care. 2011;34(5):1171-1176.

10. Li Y, Gao X, Winkelman JW, et al. Association between sleeping difficulty and type 2 diabetes

in women. Diabetologia. 2016:719-727.

11. Lin D, Sun K, Li F, et al. Association between habitual daytime napping and metabolic

syndrome: a population-based study. Metabolism. 2014;63(12):1520-1527.

12. Wu J, Xu G, Shen L, et al. Daily sleep duration and risk of metabolic syndrome among

middle-aged and older Chinese adults: cross-sectional evidence from the Dongfeng–Tongji cohort study. BMC Public Health. 2015;15(1):178.

13. Gonnissen HKJ, Hursel R, Rutters F, Martens EAP, Westerterp-Plantenga MS. Effects of sleep

fragmentation on appetite and related hormone concentrations over 24 h in healthy men. Br J Nutr. 2013;109(4):748-756.

14. Huerta C, Abbing-Karahagopian V, Requena G, et al. Exposure to benzodiazepines

(anxiolytics, hypnotics and related drugs) in seven European electronic healthcare databases: a cross-national descriptive study from the PROTECT-EU Project. Pharmacoepidemiol Drug Saf. 2016;25 Suppl 1:56-65.

15. NHG-Standaard Slaapproblemen en slaapmiddelen | NHG.

https://www.nhg.org/standaarden/volledig/nhg-standaard-slaapproblemen-en-slaapmiddelen. Accessed May 29, 2016.

16. Expanded CI, Index C, Ci- CP. Antidepressant Medication as a Risk. 2013;36.

17. Hägg M, Houston B, Elmståhl S, Ekström H, Wann-Hansson C. Sleep quality, use of hypnotics

(18)

107

18. Jakubowski KP, Hall MH, Lee L, Matthews KA. Temporal Relationships between Napping and

Nocturnal Sleep in Healthy Adolescents. Behav Sleep Med. 2016;2002(April):1-13.

19. van ’t Riet E, Alssema M, Rijkelijkhuizen JM, Kostense PJ, Nijpels G, Dekker JM. Relationship

between A1C and glucose levels in the general Dutch population: the new Hoorn study. Diabetes Care. 2010;33(1):61-66.

20. Sleep questionnaire_24-10-2012.

21. Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index:

a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2):193-213.

22. Expert Panel on Detection E and T of HBC in A. Executive Summary of the Third Report of

the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA J Am Med Assoc. 2001;285(19):2486-2497.

23. Seidell JC, Oosterlee A, Deurenberg P, Hautvast JG, Ruijs JH. Abdominal fat depots measured

with computed tomography: effects of degree of obesity, sex, and age. Eur J Clin Nutr. 1988;42(9):805-815. http://www.ncbi.nlm.nih.gov/pubmed/3181112. Accessed May 9, 2016.

24. WHO | Physical Activity and Adults. WHO. 2015.

25. CESD-R: Center for Epidemiologic Studies Depression Scale Revised Online Depression

Assessment » CESD-R Explanation. http://cesd-r.com/cesdr/. Accessed May 5, 2016.

26. Overestimation of risk ratios by odds ratios in trials and cohort studies: alternatives to

logistic regression. 2012;184(8):895-899.

27. Ju S-Y, Choi W-S. Sleep duration and metabolic syndrome in adult populations: a

meta-analysis of observational studies. Nutr Diabetes. 2013;3(5):e65.

28. Matthews KA, Pantesco EJM. Sleep characteristics and cardiovascular risk in children and

adolescents: An enumerative review. Sleep Med. 2014;18:36-49.

29. Lakerveld J, Mackenbach JD, Horvath E, et al. The relation between sleep duration and

sedentary behaviours in European adults. Obes Rev. 2016;17(S1):62-67.

(19)